-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Helper function to initialize LIF state #156
Conversation
Test failing due to pytest upgrade while old tests are using older version of nengo. nengo/nengo@9e8f363#diff-b5e89441c0e97432af7158d317237833. Thinking I should use an older version of pytest for the Nengo versions prior to nengo/nengo#1443. |
6e92457
to
dfc5e79
Compare
Codecov Report
@@ Coverage Diff @@
## master #156 +/- ##
=====================================
Coverage 100% 100%
=====================================
Files 29 30 +1
Lines 1398 1431 +33
Branches 162 165 +3
=====================================
+ Hits 1398 1431 +33
Continue to review full report at Codecov.
|
Closes issue #150.
This fixes the issue in nengo/nengo#1122 which was triggered by a few tests in python2.7
b97a23e
to
c529239
Compare
Related to nengo/nengo#1415.
This adds a helper function that can be used as follows:
to initialize the state of the
LIF
ensemblex
by sampling the steady-state solution given the zero vector as input. This is mathematically equivalent to fast-forwarding in time (to after the mixing time of the stochastic process, and after any subthreshold neurons have stabilized atv = J
), which solves the problem of the start of the simulation having some initial transient / degenerate state.It also takes into account the tuning curves (e.g., for thresholding ensembles), and supports initializing at values other than 0, including higher-dimensions.
TODO: